Recalculating Design Flood Values under Nonstationary Conditions in the Yalong River Basin, China

Résumé fourni par la source

Global climate change and land use change are the two main factors influencing the hydrological cycle. Under future climate change conditions, the consequential change of hydrological extremes and design flood values of basins will directly influence the operation and dispatch of cascade reservoirs. Therefore, studying the impact of climate change on design flood values is crucial for flood defense and hydropower output increment. Annual maximum (AM) sampling was employed to obtain the annual maximum daily flood series of the Xiaodeshi station in the future (up to the year 2100) based on soil and water assessment tool (SWAT) modeling of daily runoff. Then, a time-varying moment model with time and meteorological factors taken as the covariates was developed, and the parameters were optimized based on the particle swarm optimization (PSO) algorithm. Subsequently, the design flood values with different frequencies were calculated, varying annually from 2022 to 2100. To align with the present reservoir operation protocol, a new method was developed based on the equivalent reliability (ER) principle to calculate the unique design flood values based on the dynamic outputs. The results indicate that, compared with using time as the covariate, the time-varying moment model with meteorological factors as covariates is physically more significant and yields more reasonable results. Under future climate change conditions, the design values decrease slightly under the representative concentration pathways (RCP)2.6 and RCP4.5 concentration path for the low frequencies (exceedance probability of 0.1% and 0.2%), with increases for other frequencies (Yan et al. 2017). They increase under the RCP8.5 concentration paths for all, especially medium, frequencies (exceedance probability of 5% and 10%). By combining the time-varying moment model with the equal reliability method, more reliable design flood values can be obtained.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.